304 
NATURE 
[May 23, 1912 
optical activity. | Nierenstein has supposed tannin 
to be a mixture of digallic acid and optically active 
leucotannin; this does not agree with the slight 
acidity of tannin. The recently published researches 
of Emil Fischer in conjunction with Freudenberg 
throw a new light on the question. The authors show 
that carefully purified tannin contains about 8 per 
cent. of glucose in its molecule. They do not regard 
tannin as a glucoside of gallic acid, but consider 
that it is an acyl compound of glucose analogous 
to the penta-acetyl and penta-benzoyl derivatives of 
this sugar, in which the alcohol groups form esters 
with the acid. This novel conception of tannin as 
a penta-digalloyl glucose is in agreement with its 
chemical behaviour, but, as is his custom, Fischer 
has had recourse to synthesis to confirm his views. 
Digallic acid was not available, but a synthetic penta- 
galloyl glucose could be obtained without great diffi- 
culty, it sufficing to combine glucose with tricarbo- 
methoxygalloyl chloride in presence of quinoline and 
remove the  tricarbomethoxy-groups by cautious 
hydrolysis with alkali. The new compound has all 
the properties of the tannins, and there can be little 
doubt that the new conception is the correct one, and 
that synthetic tannin will shortly be added to the 
achievements of the organic chemist. 
SuccEssFuL trials have just been concluded of the 
first Clyde-built motor ship, Jutlandia. An _ illus- 
trated article in The Engineer for May 17 gives some 
particulars of this ship, which is a sister ship to the 
Selandia. The builders, Messrs. Barclay, Curle and 
Co., Ltd., Whiteinch, have fitted Diesel engines of 
the four-cycle type. On the measured mile at Skel- 
morlie the vessel attained a mean speed of 12 knots, 
the engines developing 2700 indicated horse-power at 
135 revolutions per minute. The fuel used on the 
trials was oil of specific gravity 0-855, but the engines 
are capable of using heavier oil, such as is obtain- 
able from the Roumanian or American oil fields. 
The builders estimate that in regular service the 
quantity of fuel necessary will be about 10 tons per 
day of twenty-four hours’ continuous running. 
Engineering for May 17 gives an account of the 
system of ozone production and distributing plant 
installed for air purification and ventilation on the 
Central London Railway by the firm of Ozonair, Ltd. 
The system is a plenum one, and consists of Sirocco 
fans placed at each of the underground stations, 
excepting that at Shepherd’s Bush. The total air 
supplied to the tubes is about 80,000,000 cubic feet 
per day. Each fan draws its air through a filter 
screen, and works in conjunction with an ozone- 
generating plant. The latter consists of mica sheets 
with metallic gauze on each side, stacked side by side, 
and energised by alternating current at about 5000 
volts, in such connection that a silent discharge passes 
between the various plates, so that air flowing 
between them is ozonised. The ozone generator is 
supplied from a small transformer, which in turn is 
supplied with 380 volts alternating current from a 
small rotary converter. The converter is connected 
on its direct-current side to the 550-volt lighting 
circuit of the railway. 
NO. 2221, VOL. 89] 
OUR ASTRONOMICAL COLUMN. 
Tue Recenr Soiar Eciipse.—Many preliminary 
accounts of the solar eclipse of April 17 are given in 
the May number of L’Astronomie; the June number 
is to be devoted to a fuller discussion. 
M. Flammarion—who is to receive the Cross of the 
Legion of Honour—gives, inter alia, a map on which 
he has drawn the central line derived from numerous 
observations. Going from south-west to north-east, 
it passes very slightly to the north of St. Nom-la- 
Bretéche, rather further north of St. Germain-en- 
Laye, south of Maisons-Laffite, north of Sartrouville, 
over Franconville and Moisselles, north of Villiers- 
la-Sec and south of Luzarches. 
Analysing the observations made from a dirigible 
near the last-named place, Comte de la Baume 
Pluvinel finds that the line is 1-8 km. north of that 
given by the Connaissance des Temps, and that 
the central phase occurred at the time given by the 
‘American Ephemeris’’; this was from fifteen to 
twenty-five seconds earlier than the times given by 
other ephemerides. The shadow of the moon, as seen 
from the dirigible, appeared as a greyish circle 3-5 km. 
in diameter travelling over the ground at about 800 
metres per second. This shadow passed over the 
villages Belloy and _ Villiers-la-Sec at the same 
moment, the former lying near its northern limit. 
At Sartrouville, M. Tramblay determined the 
interval between the appearance and disappearance of 
the cusps as four seconds. M. G. Renaudot, at 
Paris, made some very definite and _ interesting 
observations of the effects on birds and certain plants, 
which in every case behaved as they usually do at 
nightfall. As the eclipse was neither total nor 
annular, M. Flammarion suggests the designation 
d’éclipse perlée, which would describe the appearance 
of a collar of irregular pearls seen at maximum phase. 
M. Simonin asks that observers will forward to 
him, at the Paris Observatory, the results of their 
observations of this eclipse. 
SoLaR PROMINENCES IN 1910.—We have received 
amended tables of Prof. Riccd’s summary of the 
prominences observed during 1910 at Catania, in 
which some of the values are essentially different from 
those previously given, which we briefly noted in 
these columns on May 9. The mean frequencies for 
the four trimestres should read :—N. hemisphere, 1-9, 
1-5, I-I, and 0:3; S. hemisphere, 1-7, 1-3, 1:2, and 1-4, 
the mean frequencies for the year being N. 1-2 and 
S. 14. Compared with 1909, the year showed a 
decrease in the frequency and the size of the promin- 
ences. Considering their distribution, there were two 
principal maxima at 25°-29° and 55°-59°, respectively, 
in the northern, and two at 15°-19° and 50°-54°, re- 
spectively, in the southern, hemisphere. 
Tue Unirep States NavaL OssrerRvATory.—The 
superintendent’s report of the work performed at the 
U.S. Naval Observatory for the year ending June 30, 
IgII, contains, among many other items, several 
interesting results of investigations of instrumental 
errors. A wide difference of opinion among the staff 
concerning the performance of the 6-in. transit circle 
has been settled, as the result of an investigation 
lasting over three years, by a declaration that the 
instrument is fit for the fundamental observations for 
which it is now to be employed. Another investiga- 
tion was carried out to determine the cause of a 
periodic error of exactly four minutes, having a range 
of more than 5 seconds of arc, in the driving-clock of 
the 26-in. equatorial. No single cause could be 
found, so it was decided to correct the error by intro- 
ducing one of opposite sign and having the same 
amplitude. This was done by scraping the driving 
